[Tutor] Readlines

[Don Arnold]

----- Original Message -----
From: "Adam Vardy" <anvardy@roadrunner.nf.net>
To: <tutor@python.org>
Sent: Saturday, June 21, 2003 2:49 PM
Subject: Re: [Tutor] Readlines


>
> Now I'm trying Read(). I have a couple observations, and wonder what
> you folks think. Take the code below. I notice that Read() is
> apparently incredibly slow! Text appears on screen, cursor slowly
> gliding across. Which surprises as it is only reading a byte at a
> time. On a Gigahertz machine.
>
> The other thing is I notice this extract does not run corrently. It's
> just supposed to print what it reads. And a part to pause for a
> moment, after its printed many lines. (Thought that would be
> necessary!)
>
> You'll probably notice the error quicker than I have. I was just
> wondering, how can you prevent errors. Suppose its just too warm in
> the summer to pick out things easily. Well, would other ways to
> structure your code occur to you that may minimize the liklihood of
> such errors?
>
> #
> fhand=open(filepath,'r')
>
> while c!='':
>     while z!='\n':
>         z=fhand.read(1)
>         print z,
>     ln=ln+1
>     c=z
>     if ln%100==0:
>         print ln
>         time.sleep(1)
>
> print "DONE."
>

Well, since filepath, c, z, and ln aren't defined (and time wasn't
imported), this script doesn't run. If you do define them and add the
import, you still have problems: as soon as you hit your first '\n', you
stop reading from the file and just keep looping, pausing 1 second every 100
iterations. File processing like this is usually done by performing a
priming read to set up your looping condition, then having another read at
the end of your loop body:

import time

filepath = 'c:/temp2/infile.txt'

fhand=open(filepath,'r')

ln = 0

z = fhand.read(1)

while z != '':
    if z == '\n':
        ln = ln + 1
        if ln % 2 == 0:
            print '[%s]' % ln
            time.sleep(1)
        else:
            print
    else:
        print z,

    z=fhand.read(1)

fhand.close()
print "DONE."

>>>
l i n e   1
l i n e   2 [2]
l i n e   3
l i n e   4 [4]
l i n e   5
l i n e   6 [6]
l i n e   7
l i n e   8 [8]
l i n e   9
l i n e   1 0 [10]
l i n e   1 1
l i n e   1 2 [12]
l i n e   1 3
l i n e   1 4 [14]
l i n e   1 5
l i n e   1 6 [16]
l i n e   1 7
l i n e   1 8 [18]
l i n e   1 9
l i n e   2 0 [20]

DONE.
>>>

I'm still not sure why you're not just using readline( ), though:

import time

filepath = 'c:/temp2/infile.txt'
fhand=open(filepath,'r')
z = fhand.readline()

ln = 0

while z != '':
    print z[:-1],      ## don't print trailing newline since we might want
to print the line #
    ln = ln + 1
    if ln % 2 == 0:
        print ' [%s]' % ln
        time.sleep(1)
    else:
        print
    z=fhand.readline()

fhand.close()
print "DONE."


HTH,
Don